Engrava runs on your own machine — the client starts it, so there is no endpoint to ping. 151 installs a week from pypi. Last commit 11 Aug 2026.
Tamper-evident audit, MindQL, and Action Records — embedded memory database for AI agents.
Today is the operative word: we check Engrava every 15 minutes and re-read its code on every release. Watch it and you find out the day that stops being true.
This server runs on your own machine — install it with the package manager and the client starts it for you. Package name taken from the official registry entry.
claude mcp add engrava -- uvx engrava-mcp
{
"mcpServers": {
"engrava": {
"args": [
"engrava-mcp"
],
"command": "uvx"
}
}
}
[mcp_servers.engrava]
command = "uvx"
args = ["engrava-mcp"]
{
"mcpServers": {
"engrava": {
"args": [
"engrava-mcp"
],
"command": "uvx"
}
}
}
{
"mcpServers": {
"engrava": {
"args": [
"engrava-mcp"
],
"command": "uvx"
}
}
}
Auditable memory for coding agents: evidence chains, supersession history, keyless embedded SQLite
Markdown memory for AI agents. Files you can read, edit, grep, and commit. Not a database.
Encrypted-first embedded database with vector search and agent memory, exposed as MCP tools
Durable causal memory for AI agents — semantic recall, causal lineage, and skills. Local-first.
Embedded memory for AI agents with SQLite, pgvector, and Neo4j graph search.
Embedded, local-first memory and retrieval for AI agents. One SQLite file, no server, no API key.
Embedded graph database for agents: declare association rules once, edges maintain themselves.
Persistent memory for AI coding agents: one Rust binary, one local SQLite database.
Answers built from our own checks of this server.